Massive Scale Cross-Border Amazon Arbitrage

[GoldFibre]

Today's Stats​

Weekly Goals​

• DONE Solve working capital issues
• PARTIAL Speed up new listing processor (add more API keys)
• PARTIAL Add more remote working features to web app
• PARTIAL Unblock Amazon Ads accounts (carryover from last week)

I was busy with some real estate things today and was almost totally occupied.

I did have an online meeting with my account manager where I think we really built up some rapport. It seems that among my competitors I'm already in the middle of the pack in terms of revenue, and generally doing better in terms of fulfillment/account health/customer feedback.

It was a slow day for sales because I lost my buy box eligibility. I suspect this is because we had some late consols and probably a couple of late deliveries in the past few days.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[GoldFibre]

Today's Stats​

Weekly Goals​

• DONE Solve working capital issues
• PARTIAL Speed up new listing processor (add more API keys)
• PARTIAL Add more remote working features to web app
• PARTIAL Unblock Amazon Ads accounts (carryover from last week)

Buy box eligibility came back sometime today although sales didn't go up too much.

Again I was dealing with real estate issues today so I only did my daily transactional work on this project: procurement, tracking, account health, etc.

 

[GoldFibre]

Today's Stats​

Weekly Goals​

• DONE Solve working capital issues
• PARTIAL Speed up new listing processor (add more API keys)
• PARTIAL Add more remote working features to web app
• PARTIAL Unblock Amazon Ads accounts (carryover from last week)

I'm traveling again this weekend. Since I have improved my remote working setup I'm hoping that I will have less manual work than last week even though my main machine is still at home.

I paused adding offers to country 2 because I'm getting too many violations and want to see how my sales do when my account is in good standing. I'm still going full speed on country 1 though since I rarely have account health issues there.

I'm also thinking I need to change up my schedule, because I am really a morning person and am only working on this at night anymore. I want to start waking up early and getting at least an hour in each morning before starting my regular day. I can still do my other transactional stuff at night because it doesn't require too much concentration.

 

[GoldFibre]

Weekly Update​

I'm happy I met the profit from the previous week despite Black Friday being over and also my account getting buybox restricted during part of the week.

My goal in the coming week is to expand to country 3. The logistics setup will be somewhat different and more expensive, but with shorter transit times. I'll use the same partner, but instead of shipping in bulk, these will be handed off to an international courier in the US.

This is kind of an experiment to see if this selling model will work elsewhere. If it does work, then I would probably invest time in a partner that can ship in bulk, or maybe have my existing partner do it for me, then find just a local partner to do the repackaging and dispatch.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open Country 3
  • WIP Register and verify w/ local Amazon marketplace
  • WIP Get pricing and SOP set up w/ existing logistics partner
  • NOT STARTED Configure bot to handle third country
  • NOT STARTED Upload at least 25k offers

I tried to register my existing entity with the local Amazon marketplace but need some additional documentation to complete the process before I can sell. So I tried to get this together today and couldn't finish it. I have a second idea that I can look into tomorrow as well. I'll pursue a few different avenues and go with whatever turns out to be the fastest way.

I messaged my logistics partner and asked for their pricing with the international courier, but didn't hear back yet. I'm going to have to follow up daily on this, but I have some idea of the cost.

I didn't start the coding, but I don't think it will be much effort to add any new country since I already have two working already.

 

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open Country 3
  • WIP Register and verify w/ local Amazon marketplace
  • WIP Get pricing and SOP set up w/ existing logistics partner
  • WIP Configure bot to handle third country
  • NOT STARTED Upload at least 25k offers

I started adding the configurations for country 3 to the bot. As part of that I had to look up the tax laws on courier imports. I found some information but definitely need to double check with a reputable source so I don't get surprises down the line.

I also have an idea for the upload - I can create a flat file from my existing listings and upload it through the seller portal to get my catalog started.

Tomorrow my plan is to wake up early and get about two hours of hard work in before starting my regular day.

 

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open Country 3
  • WIP Register and verify w/ local Amazon marketplace
  • WIP Get pricing and SOP set up w/ existing logistics partner
  • WIP Configure bot to handle third country
  • NOT STARTED Upload at least 25k offers

I woke up at 5 am this morning and got cracking.

I discovered I needed even more documents for the new country. I had two of them but not the third, so I emailed the company that helped me register my legal entity and asked them how to get it.

Then I looked up what my transit times should be to the entire region and set up my shipping settings. Since the handover to courier will happen in the US, my handling time will be much lower, but the transit time will be longer. Overall the time to delivery will be shorter than in my existing countries.

Amazon requires you to communicate in the local language of each marketplace or if there are multiple languages, in the preferred language of each customer. In country 3 there is a new language and so I translated my one automated message I send to buyers asking for a good rating and then wrote the logic to choose the right language.

I downloaded a listing upload template and filled it in with a test product, but when I tried to upload it there was an error because my account is not yet active.

Next I worked on a script to take my existing offers, recalculate the pricing & transit time for country 3, and then spit out a TSV upload file. I wasn't able to finish it this morning but should have it by tomorrow.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open Country 3
  • WIP Register and verify w/ local Amazon marketplace
  • WIP Get pricing and SOP set up w/ existing logistics partner
  • WIP Configure bot to handle third country
  • NOT STARTED Upload at least 25k offers

I set my alarm for 5 am, but had my sleep interrupted several times, so rather than be a zombie for the entire day with only two hours of sleep, I was instead a half-zombie on only four hours.

 

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open Country 3
  • WIP Register and verify w/ local Amazon marketplace
  • WIP Get pricing and SOP set up w/ existing logistics partner
  • WIP Configure bot to handle third country
  • NOT STARTED Upload at least 25k offers

Two days in a row my sleep got messed up and I didn't get up at 5am. Will try again tonight.

My bot wasn't running for most of the day because somehow I got an order on an ASIN for which I had no information in my DB. I suspect that Amazon created a new ASIN or mapped one that I knew about to another one. Anyway I finally got home at night and fixed the issue and put the bot back to work.

I tried to get the documentation I need to complete the country 3 registration but was told I need to wait until the end of the month for the document. I will brainstorm a bit more tomorrow for other ways.

 

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open Country 3
  • WIP Register and verify w/ local Amazon marketplace
  • WIP Get pricing and SOP set up w/ existing logistics partner
  • WIP Configure bot to handle third country
  • NOT STARTED Upload at least 25k offers

I got up at 5 am to get in two hours of uninterrupted work.

First I looked into one crash I had with my automated ratings request. But it looks like it was just randomness because the script worked when I ran it again.

Then I looked at the requirements for country 3 verification again. For one of the documents I realized I could change the address and I had something that worked. For the other missing document I had emailed the company that helped establish my legal entity but he hasn't replied. So I sent a follow up email to him and planned to call him today.

Since there wasn't anything else I could do on the documents in the early morning, I then worked on my upload script. I finished it but cannot test the output until my account is activated.

In the afternoon I called the guy about the document, but he wasn't sure it's possible to get something that meets the requirements. So I just uploaded my company registration to Amazon and submitted it. If it gets rejected then I will look for an alternative document.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open Country 3
  • WIP Register and verify w/ local Amazon marketplace
  • WIP Get pricing and SOP set up w/ existing logistics partner
  • WIP Configure bot to handle third country
  • NOT STARTED Upload at least 25k offers

My application for country 3 got rejected and in typical Amazon style, no reason is given.

On the seller portal there is no resubmit option - the account is basically dead in that country.

I am going to look through my network for people that work in Amazon in that region and see if I can reach out to someone to help me get the application completed.

I also started an application for a different country that I think should have less strict requirements.

 

[GoldFibre]

Weekly Update​

The business continues to grow on its own at a pretty good pace.

My goal this past week was to launch country 3, but my application got rejected on the last day. I will look for other ways to get this done as well as apply to other countries. I did modify my bot to handle a wider variety of countries, although a few more modifications are required to have it working perfectly.

So my goal for the coming week is to launch any new country.

 

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open any third country

I looked into one of the countries and they have some third party payment service that lets you use your local bank account for transfers. I guess that this service takes some cut on the currency conversion.

I signed up for that and it looks like I can already list products.

Tomorrow morning I will modify my script to work with this country and then I'll confirm the shipping rate with my vendor in the afternoon and try to launch with maybe 10k products.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open any third country

I woke up at 5 am and got some focused work done.

First I adjusted the shipping settings in the new country. Then I added the configurations in my bot for calculating handling time and price.

After generating an upload TSV, I benchmarked against the market. Most of the products I want to list were not present yet, which is a good sign. I did find a few other sellers using my model but their prices were not good and their selection was less than 100k products. Amazon US is also selling there, and I still can't compete with their pricing. But overall it looks like there is a possibility.

As a test I uploaded one spreadsheet with about 400 products in it. There were some issues with the upload and I started troubleshooting them to try to get it right.

I created one listing manually through the UI, so technically the store is launched, but I want to have at least 10k products up this week.

Then I applied for yet another country and got stopped for some documents, one of which I don't have. I've got 90 days to upload and verify the documents before they permanently close the account.

 

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open any third country

I woke up at 6 am and got in about an hour of work.

I started refactoring all my pricing code because it was a bit of a mess. Now I have country-specific config objects that colocate all of their settings and calculations. I also added a lot of tests to verify the pricing. In this process I found some bugs with my previous code, though nothing too worrying.

We also fulfilled 74 orders today, which required some coordination with the warehouse.

I also worked on some side programming project which is probably a waste of time, but I wasn't feeling 100% today.

I'm pretty sure I've caught COVID, and although it isn't severe, I'm extremely sleepy. So I'm going to nod off very early tonight and catch up on procurement in the morning.

 

[xShepherdx]

@GoldFibre this is one of my favorite threads to check in on because of your relentless execution and how open you are about your challenges and goals.

I'm not sure I even understand your business but I've learned a ton just by seeing how you respond to challenges and set goals for yourself. You're a huge inspiration, keep it up!!

And that sucks about the possible COVID. I hope it doesn't hit you too hard and you're able to get back to 100% quickly!

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[GoldFibre]

@GoldFibre this is one of my favorite threads to check in on because of your relentless execution and how open you are about your challenges and goals.

I'm not sure I even understand your business but I've learned a ton just by seeing how you respond to challenges and set goals for yourself. You're a huge inspiration, keep it up!!

And that sucks about the possible COVID. I hope it doesn't hit you too hard and you're able to get back to 100% quickly!

Thanks, I appreciate it.

People I meet in person also don't seem to understand the business after I explain it!

 

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open any third country

I took a day off from my full time job because I felt quite sick this morning.

So I actually spent a lot more time focused on this business.

Revenue Bug​

I found that sometimes my recorded revenue on my orders is much lower than what is actually found in the Amazon seller portal. So I have overstated my profit to a certain extent.

I fixed the bug in the sense that now the revenue on the order will reflect what Amazon says the revenue is. But I still didn't know why the mismatch was there. Maybe not all of my pricing updates are going through or maybe I'm now updating my DB the right way when I do update the prices.

Credit Cards​

A few days ago I started using my second Amazon credit card that has a higher limit than my business AMEX. I checked my account and the points are accumulating even though the account I buy the stuff with is not the same account on the CC itself.

I called the bank and shifted as much credit limit as possible to the Amazon card. Now between the two Amazon cards I have over 100k USD in credit that should handle my procurement even if I double or triple in sales.

New Countries​

One of the countries deactivated my account because I'm missing some documents. I called seller support and asked them what exactly has to be on the docs. The business address in my seller info has to match the address on a utility bill that I upload. I told them I pay the utility bill personally so my business name won't be on the document and they told me it shouldn't be a problem. Just to be safe though I switched my business info address to my private residence and also put my name on it. Then I uploaded the document - let's see if Amazon is OK with it.

I chased my vendor but he is traveling today and I think was in the plane when I tried to call him. So I messaged him to push him for the rates. This vendor helps me a lot but is very hard to pin down sometimes.

 

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open any third country

Not a great day.

No response from my vendor after he left on vacation.

My documents for another country got rejected by Amazon so I need to call them and ask if they can give me more information on why they were rejected.

I had a call with my account manager and got nothing out of it. No progress on any of my requests and no useful information for me. And later today I got my bill for the last three months of 'service'.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open any third country

Was busy with the full time job as well as some real estate issues pretty much the entire day.

Still getting all my procurement and customer service and such done in under an hour before bed each night though.

 

[GoldFibre]

Today's Stats​

Weekly Goals​

• Open any third country

In the back of my head I am sure that the best way to multiply this business is to run my own online store. However I was worried about the integration work involved and kept hesitating to do anything.

So this morning I just went ahead and opened a store and realized there is probably a Shopify app that does most of what I need or can at least get me started quickly. I installed marketplace connect and connected it to my Amazon account. Then I waited a REALLY long time for my listings to fetch:


While waiting I configured the store with some basic setup. My goal was to launch something by today, regardless of how bad it is. It just has to have a domain, at least one product and a way to checkout and pay.

With that in mind, I kept the default theme for now and just filled in some placeholders. I registered my domain and connected it. It is tempting to work hard on the site design, but this is really not the most important thing to work on, and can suck up a lot of time. I think I just need to prove the concept with some sales and then I can invest in a professional design.

Since it was taking a long time to get my Amazon listings, I went ahead and manually created one product (something I sold today) and launched the store. So I got my MVP and now I can work to make it better.

The main thing is to populate the products. There is an API rate limit in Shopify that prevents you from creating too many product variants in a day, but I'm not sure if it applies to this app, which is developed by Shopify. If I do hit the limit I might be able to reach out to support and get the limit lifted temporarily.

Once the products are populated then I need to work in getting traffic and setting up a conversion funnel. I'm thinking that the primary traffic sources will be Google organic and Google Ads, but a friend of mine gave me an idea of starting a separate price comparison site that compares between my store and Amazon and maybe a few other stores. With this site even if I am not the lowest price and don't get the traffic, I can still earn money through Amazon's affiliate program. Creating this site would take some time, so I'll start out with the easy setup of the Google channels.

So next I set up Google Analytics and Google Ads. There was no point in running any ads until my products are populated though, so I worked on transactional things for the rest of the morning while waiting for the Amazon listings to load in Shopify.

And then I realized my way of importing the amazon products probably wasn't going to import the pictures. So I evaluated a bunch of other apps that offer that. Most of them had some charge that scaled with the number of products, which doesn't work for me. I need to keep looking to find something that will work and in the worst case figure out the automation myself.

 

[GoldFibre]

Weekly Update​

There was a dip in orders and total profit this week. I suppose it could be that we are past the peak Christmas shopping time. In any case my bot continues to add listings, so the long term trend should continue to go up.

I failed in my goal of launching a new marketplace and kind of floundered around during the week. One thing I did do is set up a basic Shopify store as a second sales channel, although it only has a few products on it right now.

For my lack of productivity, I have the lame excuse that I got sick and barely slept, but it's still a lame excuse. If there had been a gun to my head I could have found a way to get it done.

I also got into some unproductive, time-wasting habits of watching YouTube and playing chess. This week I'm going to put myself on a complete ban of these things and report out on whether or not I held up the ban each day.

As for what to accomplish in the coming week, I want to find a way to get a large number of products onto the Shopify store, including images, descriptions and prices. This product sync doesn't have to be the final solution, just something to get started. And then after that I want to set up some Google Ad campaigns manually to start pushing some traffic.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[BizyDad]

Weekly Update​

View attachment 52884
The business continues to grow on its own at a pretty good pace.

My goal this past week was to launch country 3, but my application got rejected on the last day. I will look for other ways to get this done as well as apply to other countries. I did modify my bot to handle a wider variety of countries, although a few more modifications are required to have it working perfectly.

So my goal for the coming week is to launch any new country.

Weekly Update​

View attachment 53032

There was a dip in orders and total profit this week. I suppose it could be that we are past the peak Christmas shopping time. In any case my bot continues to add listings, so the long term trend should continue to go up.

I failed in my goal of launching a new marketplace and kind of floundered around during the week. One thing I did do is set up a basic Shopify store as a second sales channel, although it only has a few products on it right now.

For my lack of productivity, I have the lame excuse that I got sick and barely slept, but it's still a lame excuse. If there had been a gun to my head I could have found a way to get it done.

I also got into some unproductive, time-wasting habits of watching YouTube and playing chess. This week I'm going to put myself on a complete ban of these things and report out on whether or not I held up the ban each day.

As for what to accomplish in the coming week, I want to find a way to get a large number of products onto the Shopify store, including images, descriptions and prices. This product sync doesn't have to be the final solution, just something to get started. And then after that I want to set up some Google Ad campaigns manually to start pushing some traffic.

Boy these updates are starting to get reeeeally interesting.

Look at these growth figures. Out of curiosity, how much of this is seasonal buying because of Christmas?

 

[GoldFibre]

Out of curiosity, how much of this is seasonal buying because of Christmas?

It's hard to tell what is meant as a gift or not unless the customer tells me, which only a few have. In general I think orders went up for me around black Friday because of all the sales, even if I don't run any myself.

I suppose we will see a 'normalized' revenue again in January.

 

[Andy Black]

after that I want to set up some Google Ad campaigns manually to start pushing some traffic.

Following...

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[GoldFibre]

Today's Stats​

Update​

• Avoided watching YouTube - YES
• Avoided playing chess - YES
• Get lots of products on Shopify - IN PROGRESS
• Set up initial Google Ads campaigns - NO

After what feels like months, I did some real coding today. I built a script that picks ASINS from my DB, uses Puppeteer to navigate into the Shopify admin, and uses the one free Amazon importer app I found to import the product and it's variants.

It took some time to handle the iframes and the edge cases, but I got it working.

Sometimes the app complains that I've imported too many products and says I need to wait one hour, but then it starts working again after a few minutes. This definitely slows down the process. I'm pretty sure this is using the Amazon Affiliate API behind the scenes to pull the data. I had investigated this before and it is also rate limited pretty strictly.

Anyway I added a loop that will keep retrying once per minute whenever this rate limit comes up.

I will let this run on and on until I have a better/faster way to import.

 

[GoldFibre]

Today's Stats​

Update​

• Avoided watching YouTube - YES
• Avoided playing chess - YES
• Get lots of products on Shopify - IN PROGRESS
• Set up initial Google Ads campaigns - NO

By chance I woke up quite early at 4am and got to work.

I let the script run overnight, but it crashed due to some randomly appearing modal. I think this modal probably was asking for an app review. I added a fallback to reload the page and restart on any unanticipated crash to try to handle this.

Since I don't have a way to speed up the number of products I can upload per hour, I tried to think of another optimization to make. So I modified the script to prioritize uploading products that I've already sold on Amazon, rather than just anything that is in my catalog.

One thing about this import is that it is using the affiliate API and therefore it doesn't know which seller account is mine. This means that the price it imports is whichever one is winning the BuyBox, even if it isn't me, which has a risk of selling at a loss. So at some point I should have a script that automates the price/availability updates in Shopify, and maybe modify this script to immediately check and upload the correct prices after upload.

Also it isn't necessarily the case that I want the prices to be the same. When I sell on Amazon they are usually taking a 15% fee, but when I sell on Shopify I'm looking at maybe 3% for the credit card processor. On the other hand, I will have to cover the cost of whatever marketing I do.

A bigger issue is that there aren't any collections applied to the products after importing, and collections in Shopify are what drive the product categorization and navigation. Ideally I want a big nav menu with a bunch of product categories for the user to find what they are looking for. For now they can use search, so I can just make that the most prominent thing in the design.

I removed the main nav and used some custom liquid code to put the search bar front and center in the hero section. I'm roughly copying my main competitor's design here.

I needed a background image for my hero section, so I asked ChatGPT to make it for me. Here are some of the iterations:




This was a lot of fun, but I had to stop myself and settle on something to move forward with.

After that I set up Google Search Console and did some basic SEO work on the store.

 

[GoldFibre]

Today's Stats​

Update​

• Avoided watching YouTube - YES
• Avoided playing chess - YES
• Get lots of products on Shopify - IN PROGRESS
• Set up initial Google Ads campaigns - NO

I am thinking that the most valuable asset I have now may not be my Amazon Seller account, but the database of ASINs I've accumulated over the past months.

I want to see if I can use the Amazon Associates affiliate program to add a revenue stream. I know absolutely nothing about this, so to get started I read the terms and conditions and then signed up for an account.

My current idea is to add some kind of 'check price on Amazon' link on my Shopify store pages. From the terms and conditions I didn't see this explicitly prohibited, but maybe in practice they won't approve my site. I also have a backup plan of creating a price comparison site focused on this market and using the links. Since I have millions of ASINs, I should be able to make money on this long tail.

I worked together with my vendor to improve the exception handling process since we were having trouble keeping track of a few early-canceled orders that the US team needed to take care of.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[GoldFibre]

Today's Stats​

Update​

• Avoided watching YouTube - YES
• Avoided playing chess - YES
• Get lots of products on Shopify - IN PROGRESS
• Set up initial Google Ads campaigns - NO

I read some articles on the Amazon Associates program and found that a frequent reason for banning accounts is using Amazon product images and descriptions outside the context of their affiliate program. I'm pretty sure my Shopify store would qualify since I'm also using the data to sell the products myself.

I immediately removed my domain from the affiliate program to avoid getting banned.

So instead maybe I will launch my product comparison website. This is a type of site that is explicitly allowed by the affiliate program. I'm pretty sure that an MVP would be simple to set up, and I can improve it over time. I can also link to my own Shopify store there and have it feeding multiple revenue streams.

I launched a new NextJS project and outlined the important aspects:

• Need a way to do a dynamic search without it being too slow on my db
• The sitemap management has to be excellent because there will be millions of URLs
• There should be product detail pages as well as product type list pages
• All the links need to be Amazon Associates links, which need to be generated via Puppeteer until I get API access (requires three recorded sales plus Amazon approval)

I built a simple text-only version first. I was inspired by this site and learned a number of things from studying it: Disk Prices (US)

I found it at Boring Cash Cow, which @MTF shared recently.

While building this I also realized I could borrow more from the Disk Prices idea and essentially auto-generate sites/pages like theirs across hundreds of categories. Their site checks only Amazon prices and lists from multiple Amazon marketplaces.

And the advantage I have over Disk Prices is that their site relies entirely on the severely rate-limited affiliate API, and I have my much faster scraping bot that uses the SP-API.

This is a departure from my original idea of doing a price comparison between Amazon and my own store and only focusing on one country.

Anyway this is floating around in my head and I will keep building and see where it takes me.

In the afternoon I had some time and did a front page that lists many of the product types and links to the price comparison pages. I have some idea for dynamically generating the filters based on the Amazon product type attributes.

 

[GoldFibre]

Today's Stats​

Update​

• Avoided watching YouTube - YES
• Avoided playing chess - YES
• Get lots of products on Shopify - IN PROGRESS
• Set up initial Google Ads campaigns - NO

Today I was mainly tied up with other admin stuff as I am moving house soon.

In between I worked on fulfillment and customer service, which is creeping up in my daily time allocation.